CIRCLE WITH A DOT

Another talk announcement for BSides Luxembourg!🧠 𝗧𝗔𝗟𝗞 𝗧𝗢 𝗔 𝗦𝗛𝗘𝗟𝗟: 𝗘𝗫𝗣𝗟𝗢𝗜𝗧𝗜𝗡𝗚 𝗔𝗜 𝗔𝗚𝗘𝗡𝗧𝗦 𝗜𝗡 𝗥𝗘𝗔𝗟 𝗧𝗜𝗠𝗘 – Parth Shukla AI agents are no longer just chatbots—they can execute commands, access files, and interact with real systems. But what if an attacker could control all of that… just by talking?This talk reveals a real-world vulnerability where full system command execution was achieved through natural language interaction alone. From reconnaissance to bypassing safeguards using jailbreak techniques, this session shows how AI agents can become unintended attack proxies—no exploits, no credentials, just conversation.Parth Shukla is a Senior Security Researcher specializing in AI Security and Adversarial Machine Learning, focusing on securing agentic systems and uncovering real-world vulnerabilities in LLM-driven architectures. Conference Dates: 6–8 May 2026 | 09:00–18:00 14, Porte de France, Esch-sur-Alzette, Luxembourg️ Tickets: https://2026.bsides.lu/tickets/ Schedule Link: https://pretalx.com/bsidesluxembourg-2026/schedule/ #BSidesLuxembourg2026 #AISecurity #LLM #AgenticAI #CyberSecurity #AppSec #AdversarialAI

@cigitalgem Seems to be kinda difficult to find out which ant did what, and why...

@simonroses After a month with OpenClaw, I've spent this week on Hermes Agent.Right now? I'm sticking with Hermes. It feels less fragile.

Two Claude Code releases practitioners should know about:🧠 --bare flag for CI + --channels for mobile approval relayNew flags make scripted/CI invocations leaner (no hooks, LSP, plugin sync). The --channels flag lets you approve tool calls from your phone via channel servers.🧠 Critical ~50% failure rate bug fixed in multi-schema workflowsStructuredOutput schema cache was silently causing ~50% failures in agentic pipelines using multiple schemas. Also fixes prompt cache misses, memory leaks on large JSON, and OOM on 1GB+ file edits.solomonneas.dev/intel#AI #ClaudeCode #DevTools #AgenticAI

Per-alert AI pricing makes AI a non-starter for most service providers.A hundred alerts a day per tenant, multiplied across dozens of clients, priced per token. The math always comes out wrong.For MSSPs, that's not just a cost center problem. Getting the economics wrong doesn't just hurt margins, it stops AI adoption entirely.Agentic SecOps Workspace (ASW) changes that equation. Connect through Claude Code, and AI costs shift from a variable tied to alert volume to a predictable monthly rate per analyst. One tenant spiking doesn't take your margins with it.Because LimaCharlie's API covers 100% of platform functions, the scope of what that analyst can do expands well beyond triage.Onboard new clients, write and deploy detection rules, and manage cross-tenant configurations, all in the same session.LimaCharlie doesn't charge for AI. It's part of the platform.Get started at http://limacharlie.io#mssp #secops #ai #agenticai

@itsfoss Get these 23 AI generated books and ask LLM to summarise them for you. The result can be directly prompted to an AI chat bot.

A good review of reasons insurance companies are pulling back from insuring companies that lean on generative AI. Point 4, "The main problem is not just the error, but the incentive not to see it" is especially damning: use of AI not only obscures audit trails, it sets up perverse incentives against accountability, pushing costs and risk to other parts of an organization, its customers, or society. The net result is that whatever "local" advantages AI may provide turn into downstream risk that cannot be easily accounted for. Insurance companies are (rightly) allergic to this state of affairs.Another example of how (whole)-systems thinking is very helpful for parsing the effects of technology changes like this.https://freakonometrics.hypotheses.org/89367#AI #GenAI #GenerativeAI #LLMs #AgenticAI #GPT #ChatGPT #Claude #Gemini #ActuarialScience #insurance

This nicely balanced article by @cademetz is only the tiniest tip of the iceberg. The security issues surrounding control of one agent get much more complicated when a swarm of agents (say 10,000 or even 1,000,000) is what you must consider. Time for #MLsec to face the future...not look back to adapt spent solutions.#ML #AI #AgenticAI https://www.nytimes.com/2026/03/19/technology/ai-agents-uses.html?unlocked_article_code=1.UVA.L8Pk.hlPQ4ekR49lC&smid=nytcore-android-share

#Perplexity's '#PersonalComputer' Lets #AI Agents Access Your Local Fileshttps://slashdot.org/story/26/03/12/191200/perplexitys-personal-computer-lets-ai-agents-access-your-local-files#AgenticAI #privacy #cybersecurity

Silicon Valley’s new obsession: Watching bots do their grunt workhttps://www.msn.com/en-us/money/other/silicon-valley-s-new-obsession-watching-bots-do-their-grunt-work/ar-AA1YsdAP #AI #ArtificialIntelligence #AgenticAI #PersonalBots #Claude #OpenClaw

@concertina226 @Samsung I'm sure this is having a positive impact for women in many countries...https://www.malwarebytes.com/blog/news/2025/11/budget-samsung-phones-shipped-with-unremovable-spyware-say-researchers#spyware

@kaidu Ja, das passt wie die Faust aufs Auge. Genau das ist auch das Ziel der RentAHuman Gründer. Und alle helfen ihnen mit dem Marketing, indem sie den Scam auch noch verbreiten und deren Lügen ungeprüft übernehmen.